People are aware of pronunciation, if not phonology and phonetics. Phonetics is the study of the production, transmission and reception of sounds in human speech and is the subject of careful, informed, scientific investigation. Phonology is the study of the usage of speech sounds in language in general and languages in particular and is the subject of careful, informed, linguistic investigation. Pronunciation, on the other hand, is everyone’s activity in spoken communication; that is what we hear and what we produce in talk; it is everyone’s experience and hence is part of everyone’s business.

In this chapter we begin with one accent of English – Southern England Standard Pronunciation (SESP) – although the principles of description will apply to any accent. We will confine our attention to the phonology of monomorphemic words in citation form as an initial introduction to features of Systemic Phonology and to the basic design of system network displays. System in word phonology is not like system in lexicogrammar or intonation, as sets of options from which a speaker chooses to create meaning; system at the rank of word (and also, largely, at the rank of groups/phrases) is rather the specifications of what the speakers of a language recognize as having been established in their language to represent its words. Moreover, as we have maintained, the primary function in phonology is to serve lexicogrammar by providing each distinct unit with their uniquely distinctive shapes. However, as we have noted previously, we have to concede that in the historical development of the language, phonological shapes that were once distinctive and unique may no longer be so, and ‘accidentally’ become homophones; another example is English right, write, rite, wright, which were, in Early Modern English phonology, distinct: /rɪçt/, /ʋritǝ/, /ritǝ/, /ʋrɪçt/ (see e.g. Gimson 2014: 66–73); all eventually became SESP /raɪt/ through various historical processes. Nevertheless, apart from such historical accidents, the primary function of phonology is to provide a distinctive shape to all the discrete units of the lexicogrammar. Every word in citation form must be represented by one foot, apart from the few instances indicated in 1.4. This is why in this presentation of Systemic Phonology the specifications of types of foot comes first, rather than the bottom-up approach from consonant and vowels segments. As we stated in 1.5, a full description of the phonology of words of any language would ideally include statements about the structures of the foot and its range of prosodic shapes, the permissible number of syllables (‘syllabic count’), the permissible kinds of structure in a syllable, the inventory of phonemes at the nucleus of the syllable and at the margins, permissible combinations of phonemes and specific allophonic details.

In this chapter we extend our presentation of the phonology of English words from free, basic, monomorphemic words to a consideration of words with complex structures, i.e. with affixes and compound elements. Linguists in the SFL tradition have been preoccupied with the grammar of the clause and have not shown a great deal of interest in morphology; it barely gets a mention beyond an acknowledgement that morphemes exist as the very smallest unit in the grammatical hierarchy (see Halliday et al. 1964: 51; Berry 1975: 91; Young 1980: 13–14; Halliday 1984: 20; Thompson 1996: 20; Downing and Locke 2004: 10, 13; Halliday and Matthiessen 2004: 9; Fontaine 2013: 35). For this reason, we must seek support from elsewhere, from other traditions including psycholinguists such as Aitchison (2003). We will show how the principal features of Systemic Phonology apply to complex, polymorphemic words as well as simple, free forms. Affixes, i.e. bound morphemes, provide specifications of details of a grammatical nature to the meanings of the words and stems that they are attached to. Likewise, compound elements provide a more specific focus of the meaning of the words and stems they are attached to. Just as the primary function of phonology is to provide a (more or less) unique spoken form of all the units that make up a language’s lexicogrammar, this applies to affixes and compound structures too. We noted that typically words in citation form are represented in phonology as feet, which display a prosodic shape and a structure of syllables, which themselves have a structure of phonemes. In the case of compound words in citation form, there are two elements, each represented by a foot, but one subordinate to the other as an ‘extension’; however, in a small number of cases the second element is simply an unstressed syllable. In the case of affixes, the matching phonology is the syllable and phoneme. We will deal with the bound morphemes of affixes first, where, traditionally, a division has been established between affixes that designate either inflections or derivations. Inflections indicate the addition of a grammatical category to a word that does not alter their word class designation, whereas derivations indicate a change of word class or an additional category of meaning. We consider inflections first, and then derivations, with a brief interlude on numerals that share some features of both. And then, finally, compounds.

German is a cousin of English in the family of Germanic languages; hence the many notable similarities in vocabulary, grammar and phonology. At the level of word phonology, the similarity includes a set of plosives with matching nasals, a relatively large set of fricatives but nevertheless with marked differences, the sonorants /l, r/ but with a variation in phonetic features, a rich vowel system with short and long vowels, and marking of word stress in the first syllable of the foot. These are features common to most Germanic languages. German word phonology distinguishes itself from English principally by the more restricted distribution of voiced obstruents in syllable/wordfinal position, the addition of rounded front vowels and a much smaller system of diphthongs; also there is a strong preference for keeping word stress stable in extended word forms, which allows us to attempt to handle both mono- and poly-morphemic words together.

Welsh is spoken extensively in Wales, by about 700,000 people in a total population of approximately 3 million. It is also spoken in Argentina by as many as 5,000 people as a result of Welsh emigration from the mid-nineteenth century into areas of Patagonia (BBC 2014), and also, in smaller numbers, in USA, Canada and the rest of UK, again as a result of migration. Welsh is a Celtic language, in the Brythonic branch, which also includes Breton and Cornish. It is an official language in Wales alongside English. Welsh has a number of notable features. It is a verb-subject-object (VSO) language. It has a distinctive word class of verb-nouns, some of which also operate within larger verbal groups to form the equivalent of modals and the passive. Prepositions are conjugated like verbs. Possessive pronouns consist of elements preceding and following a noun. In phonology, there is a very strong tendency to maintain penultimate stress whatever suffixes are added to a word; initial consonants are subject to change (‘mutation’) in certain syntactic contexts, including definite reference with feminine nouns; and the phonotactics of vowels and final consonants within a syllable or word is intriguingly elaborate. A sociolinguistic feature of Welsh is that there is no single standard accent in Wales. However, the collection of accents in the North of Wales have enough common factors to suggest, in general terms, a North-Walian accent (see, e.g. Kelly 1992), and similarly, a collection of accents in the South have enough common factors to suggest a South-Walian accent. Several studies show a gradation of phonetic and systemic features between Northern and Southern accents, with the South-West being the simplest system of all (see Rees 2018 and his cited bibliography). It is commonly heard that North-Walians claim that their accent is the purest because it is the richest in terms of systemic inventories. It is also the case that the percentages of Welsh speakers in the populations of the North is very much greater than those in the South; nevertheless, the populations of the South are greater in numbers. It is therefore necessary to either develop two separate system networks for the North and the South, or combine them in such a way that the South system is shown to be a reduction of the North system. The description of Welsh phonology that underlies this new presentation in terms of Systemic Phonology is based upon the following scholarly work: Awbery (1984; 2009), Ball and Jones (1984), Ball and Williams (2001), Clee (2017), Hannahs (2009; 2013), Iosad (2017), Kelly (1992), Mayr and Davies (2011), Rees (2018) and Williams (1989), as well as local informants. In the Argentinian province of Patagonia, Welsh has been spoken continuously since the first Welsh settlers landed in 1865. As more Welsh settlers arrived in the nineteenth century, their language community became well established. Although the settlers originated from all parts of Wales, it appears that elements of the South-Walian accents prevailed even though the Northern syntactic and lexical items of the Welsh Bible were preserved. Use of the language declined in the first half of twentieth century under pressure from government language policies and the inward migration of Spanish-speaking workers attracted into a relatively prosperous economic environment. However, a revitalization of the language began with the centennial celebrations of 1965, and continued with eisteddfod events, the remarkable impact of a dedicated, highly respected Welsh missionary, Mair Davies in the Welsh chapels (see Davies 2012), and the Welsh Language Project managed locally since 1997 but financed principally from Wales (Arwel 2016; see also Johnson (2009; 2010) and Coupland and Garrett (2010)). The impact of this revitalization has led to a re-evaluation of the significance of Patagonian Welsh within the Welsh language community in Wales, which justifies its inclusion in this present study.

Tera is one of the 500 indigenous languages of Nigeria. It is a minority language spoken by over 100,000 people in the north east of the country, in the northern part of Gombe State and the eastern part of Borno. Most of the people acquire Hausa as their trade language and a smaller proportion acquire English as their language of education. Their own name for themselves and their language is Nyimatli /ɲimáɬi/. The language is classified as Afro-Asiatic, Chadic, Biu-Mandara, A, A.1, Western (Eberhard et al. 2019). Linguistic work on Tera has principally been associated with Newman (1964; 1968; 1970; 1980). However, an orthographical version of the language was established much earlier with the publication by the British and Foreign Bible Society of Labar Mbarkandu nu Yohanna Bula Ki, a first translation of the Gospel of John, attributed to Gordon Beacham, a Canadian missionary, in 1930. This was accompanied by a catechism and a hymn book, but the use of Tera in Christian worship gradually gave way to the dominance of Hausa. A revival of interest in a written form of the language occurred in the 1990s with Ayuba Nyagham’s Ye Chituku ɓu Me Nyimatli, and used quite extensively by the Tera writer and broadcaster Jauro Maila. In 2004, Isioma A Jideonwo, published her Let’s Develop Nyimatli Language, and Maimuna Magaji, her Tera-English-Hausa Dictionary in 2018. In the meantime, the Nyimatli Language Project conducted a series of workshops from 2004 that included the creation of a new orthography, which resulted in the publication of Reading & Writing Nyimatli in a definitive edition (NLP 2015), along with a set of reading books, Lagarkati Shogar Me Nyimatli, and the Nigerian Bible Translation Trust published the complete New Testament, Mewar Alqawarang, in 2016. A more recent treatment of Tera pronunciation appeared in Tench (2007).